indapa has asked for the wisdom of the Perl Monks concerning the following question:

PerlMonks, I have simple regular expression question for you. I am reading a file and if a line starts with the number 1, I want to write that line to a file. The problem is that if that file starts with 1, 11, 12, etc. those lines get written that file too. My regular expressions is like this: 
if (/^1/) {
    print FILE, "$_\n";
}

[download]
Is there a way to make my regex more stringent so I don't get lines in my file I don't want? Thank you.

Replies are listed 'Best First'.
Re: simple regex question
by beowulf (Scribe) on Mar 03, 2001 at 07:31 UTC
    Assuming that you want just the lines that have 1 and then whitespace after that, and then text the following code will work

    if (/^1\s+/) {
    print FILE, "$_\n";
}

    [download]
    Life? What's a life?
[reply]
[d/l]
      A more general solution might be /^1\b/. Of course, the previous suggestion will almost certainly work also, but in the spirit of TIMTOWTDI, and not fully knowing the format of the data..

       
      perl -e 'print "I love $^X$\"$]!$/"#$&V"+@( NO CARRIER'

[reply]
[d/l]
        A more general solution might be /^1\b/.

        ...or, borrowing from Tuna below: /^1\D/.

        p
[reply]
[d/l]
[select]
        I was thinking about that too, actually
        Life? What's a life?
[reply]
      Life? What's a life?
      anagram of a file? an ordered sequence ending with EOL?
[reply]
Re: simple regex question
by Tuna (Friar) on Mar 03, 2001 at 07:41 UTC
    Here's a simple one, that seems to cover it:

    The "data" is a file called "lines":

    1 flew over the cuckoos nest
    12 flew over the cuckoos nest 
    #!/usr/bin/perl -w

use strict;

my $line = "/home/trix/lines";

open (L, "$line") || die "$!\n";
while (my $lines = <L>) {
  chomp $lines;
  if ($lines =~ /^1{1}\D/) {
    print "my lines are $lines\n";
  }
  else { print "Nope!\n";
       }
}

    [download]


    Update: this regex doesn't catch lines that begin with "1d", for instance. Take other's suggestions.
[reply]
[d/l]
Re: simple regex question
by arturo (Vicar) on Mar 03, 2001 at 23:14 UTC

    Just for fun: 

    while (<FILE>) {
    print if ( /^(\d+)/ && $1 == 1);
}

    [download]

    Oughta do the job too.

    (Explanation: if the expression matches, i.e. if there are one or more digits at the beginning of the line, then that sequence of digits will be saved in $1, and then the right-hand-side of the conjunction will be evaluated).

    Philosophy can be made out of anything. Or less -- Jerry A. Fodor

[reply]
[d/l]
[select]
Re: simple regex question
by Madams (Pilgrim) on Mar 04, 2001 at 10:14 UTC
    Tilly to the rescue....
    I had a nearly identical problem here it is and tilly's solution:
    my substitution problem

    Update:fixed the messed up link thanks a and Corion :)
    Cheers,
    _________________
    madams@scc.net 
                  (__)
              (\/)
       /-------\/
      / | 666 ||
     *  ||----||

    [download]
[reply]
[d/l]
Re: simple regex question
by indapa (Monk) on Mar 04, 2001 at 01:03 UTC
    Thanks for the responses, the following regex did the job: 
    if (/\b^1\b/) {
   print FILE, "$_\n";
}

    [download]
[reply]
[d/l]

Back to Seekers of Perl Wisdom